{"id":"W2031600655","doi":"10.1109/3pgcic.2012.38","title":"Large-Scale Mining of Co-occurrences: Challenges and Solutions","year":2012,"lang":"en","type":"article","venue":"","topic":"Data Mining Algorithms and Applications","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Computer science; Set (abstract data type); Pruning; Field (mathematics); Hash function; Scale (ratio); Database transaction; Data mining; Principal (computer security); Word (group theory); Information retrieval; Association rule learning; Core (optical fiber); Data science; Theoretical computer science; Database; Computer security; Mathematics","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009471351,0.002105092,0.006087747,0.005781461,0.003659745,0.007021959,0.008137885,0.004831902,0.001530113],"category_scores_gemma":[0.04650215,0.002390581,0.002386906,0.01644157,0.002758605,0.01210546,0.006495235,0.004585633,0.002249161],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0013319,"about_ca_system_score_gemma":0.003031754,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004289937,"about_ca_topic_score_gemma":0.004790216,"domain_scores_codex":[0.9862641,0.00397873,0.001268732,0.003634106,0.004300505,0.0005538468],"domain_scores_gemma":[0.9174751,0.05398066,0.006614036,0.0109893,0.009100981,0.001839864],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005698631,0.0008544727,0.02815232,0.002016757,0.0006752329,0.002222152,0.001707286,0.07927552,0.00713464,0.02566625,0.05233919,0.7993864],"study_design_scores_gemma":[0.0001522713,0.0001531133,0.007350957,0.0002551932,0.0001537069,0.00372821,0.004053213,0.7340338,0.006898703,0.2123596,0.03064496,0.0002164064],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04308996,0.01364168,0.9200337,0.01560062,0.0004450182,0.0003417289,0.001540037,0.003352235,0.001954861],"genre_scores_gemma":[0.1765267,0.005917482,0.8085147,0.001292968,0.001049712,0.0005050181,0.00374308,0.0004856524,0.001964589],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009471351,"threshold_uncertainty_score":0.0500899,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07059073906538048,"score_gpt":0.2994183145528519,"score_spread":0.2288275754874714,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}